Human 的Face Recognition using Line Features.docVIP

Human Face Recognition using Line Features Mrinal Kanti Bhowmik1 ,Debotosh Bhattacharjee2 , Mita Nasipuri2 , Dipak Kumar Basu 2, Mahantapas Kundu2 1Department of Computer Science and Engineering , Tripura University , Suryamaninagar- 799130,Tripura, India 2Department of Computer Science and Engineering, Jadavpur University, Kolkata- 700032, India ABSTRACT In this work we investigate a novel approach to handle the challenges of face recognition, whichincludes rotation, scale, occlusion, illumination etc. Here, we have used thermal face images as those are capable to minimize the affect of illumination changes and occlusion due to moustache,beards, adornments etc. The proposed approach registers the training and testing thermal faceimages in polar coordinate, which is capable to handle complicacies introduced by scaling androtation. Line features are extracted from thermal polar images and feature vectors are constructedusing these line. Feature vectors thus obtained passes through principal component analysis (PCA)for the dimensionality reduction of feature vectors. Finally, the images projected into eigenspaceare classified using a multi-layer perceptron. In the experiments we have used Object Trackingand Classification Beyond Visible Spectrum (OTCBVS) database. Experimental results show thatthe proposed approach significantly improves the verification and identification performance andthe success rate is 99.25%. Keywords: Face recognition, Thermal face images, Polar transformation, Principal component analysis, Line features, Multilayer neural network, Backpropagation learning, Classification INTRODUCTION To be safe and secure in the present world it is required to authenticate an authorized person, to keep suspects under constant surveillance, and also to investigate about criminals, missing and traceless individuals within a reasonable cost structure is a major problem. A security system based on human face is non-intrusive in nature, i.e. without any consent